« Future Pundit on Cloning | Gene Expression Front Page | Enlightened noblesse oblige »
January 27, 2003

New genes for IQ

From the journal Cell (one of the holy Trinity of journals...at least to biologists), we have the following article on the BDNF gene (subscription required). A juicy excerpt for our crew of bloggers:

"Our results also speak to a current debate about the genetics of human intelligence. Specific cognitive abilities are partially correlated, and the extent of this correlation is referred to as ''g'', synonymous with IQ. Much genetic epidemiological data suggest that the genetic determinants of different cognitive abilities largely overlap with each other and with IQ. In other words, genes influence specific cognitive traits by virtue of their ''top down'' effects on g (Plomin, 1999 ). The alternative ''bottom up'' view that genes contribute to IQ primarily through their unique and more direct effects on specific cognitive modules has found little support. The present data are consistent with the latter formulation. We did not detect an effect of BDNF on IQ or any other cognitive ability (although we did not examine some other types of memory, such as priming, skill and habit acquisition, and fear conditioning). A recent study of the role of the COMT gene on working memory and prefrontal physiology, but not IQ, provides further evidence for this view (Egan et al., 2001b ). While COMT and BDNF may ultimately be shown to exert small effects on IQ, our current data suggest that their effects are somewhat specific and mediated through ''modular'' cognitive elements."

So, what Egan et al are trying to say is that (in their view, and this research) genes that create brain architecture exist primarily to perform that function--in this case, allowing for memory. Effects on actual IQ are secondary (and therefore probably not selected for). Interestingly, this is one of the comments I left on Richard Bennet's blog--in my opinion, after a certain point, IQ is not really "needed" for survival advantage. If this is true, group differences in IQ may purely be due to genetic drift, sexual selection, or a combination thereof.

In the behavior department, another article on genes and aggressiveness came out.
An excerpt:

"People who are over-aggressive or excessively anxious may be missing a gene, say scientists who conducted experiments on mice. The gene, called PET-1, was removed from specially-bred mice by scientists at Case Western Reserve University in the US. They found that the mice had heightened levels of both anxiety and aggression - when the mice were given a "territory", their response time to attack an intruder was significantly lower than a normal wild mouse, and they tended to launch an attack more often. "

Again, deleting a gene is a far cry from likely naturally occuring human variants, but this research is pointing to some genetic reasons for differences in aggression levels in humans.

Oh, by the way, the original article is in Neuron (one of the holy Trinity of journals for neuroscientists)

Posted by david at 04:27 PM

Hey! This all looks interesting, but I don't follow it. Are they saying that the latest results support g, or are they saying that they support a bottom up theory.

Posted by: ManafromHeaven at January 28, 2003 08:07 PM

Bottom up theory--they say genes provide the brain architecture, and that differences in IQ are due to subtle interactions or variants among these "building block" genes.

Clear things up?


Posted by: David at January 29, 2003 10:00 AM

The Egan results to me show no more than that memory and IQ are differently determined. And if you know about idiot savants that will be no news.

Memory is in any case multidimensional. My everyday memory is a shocker. But I can remember the full name of the present Prime Minister of Japan. Can anybody else here do that?

Posted by: John Ray at January 30, 2003 05:55 PM